Special Issue: Challenges and Opportunities in Radiology
Introduction
The field of radiology is constantly advancing, as new technologies and techniques emerge to improve the quality, accuracy, and safety of imaging and interventions. Some of the current trends and challenges in radiology include:
• The application of artificial intelligence and machine learning to enhance image acquisition, reconstruction, analysis, and reporting
• The development and evaluation of novel contrast agents, tracers, and probes for molecular imaging and targeted therapy
• The optimization and standardization of radiation dose and image quality for various imaging modalities
• The integration and harmonization of imaging data from multiple sources and platforms
• The education and training of radiologists and other imaging professionals in the era of digital transformation
This special issue of the official journal of the European Society of Medicine aims to provide a comprehensive overview of the latest developments and controversies in radiology. It features original research articles, review articles, case reports, and editorials from leading experts in the field. We hope that this special issue will be informative and useful for radiologists, researchers, and students. We invite you to read the articles and share your feedback with us.
An isolated contraction of the Obliquus Capitis Inferior and Rectus Capitis Posterior Major demonstrated with Musculoskeletal Ultrasound Imaging
Abstract
Study design: A quasi-experimental study design
Background: The motion of the suboccipital region is controlled by several groups of the smaller and larger muscle group, and these might play a role in the development of cervicogenic headache. Abnormal muscle tone can lead to abnormal movement patterns resulting in altered proprioceptive information from the mechanoreceptors in this region. Thus, abnormal muscle tone will contribute to cervical dysfunction and might result in pain, joint irritation, and poor functioning of the neck. At the same time , unilateral contraction of the suboccipital muscles could lead to rotation of the atlas. This could result in a significant translation of the spinal cord toward the side of rotation within the dural sack. This homolateral shift of the spinal cord could then lead to dural tension and possibly contribute to cervicogenic headaches.
Methods: A convenience sample of 25 healthy subjects were recruited for this study. Musculoskeletal ultrasound imaging was used to measure the diameter of the Obliquus Capitis Inferior and the Rectus Capitis Posterior Major immediately following an isometric contraction into head extension and rotation.
Outcomes: An isometric-induced contraction resulted in a significant change in the diameter of the obliquus capitis inferior and contralateral rectus capitis posterior major and, thus, could affect the position of atlas in the atlantoaxial joint.
Discussion: The effect of an isometric-induced contraction of the obliquus capitis inferior and contralateral rectus capitis posterior major in a subgroup of asymptomatic individuals was measured using musculoskeletal ultrasound imaging. The results of this study indicate that the diameter of both muscles significantly changed with the isometric contraction. This study’s findings support that suboccipital muscles can contract in isolation and, thus effect the position of atlas.
A complete pipeline for glioma grading using intelligible AI on multimodal MRI data
Abstract
1) Objectives: Machine learning for binary glioma grading have been extensively used on anatomical MRI, especially using the BraTS dataset. The relevance of radiomic criteria based on multimodal imaging, including diffusion, perfusion and spectroscopy data is to be explored, as multimodal datasets are scarce, and there is no common benchmark for performance comparison.
2) Material and methods: Poitiers University Hospital provides 123 multimodal patient data. We computed 124 features and let a recursive feature elimination algorithm (RFE) yield a relevant, reduced subset of features. We trained a SVM classifier on this subset. We proposed a method to adapt the BraTS dataset to allow performance comparison with the literature. We got a performance reference point by training on anatomical data only, and showed improvements when multimodalities were added. We explored the feature relevance through the RFE subset. The RFE subset is not constant and induce variability in the performances. To smooth the variability, we applied the RFE algorithm 100 times and incremented the selected features, resulting in a global feature ranking. We also show the best classifier reached on these 100 trainings and its feature subset.
3) Results: The best classifier reached 86.5% accuracy, with a mean accuracy on 100 trainings of 78.6%. The rankings shows that anatomical and perfusion sequences are the most relevant for glioma grading, especially T1 post-gadolinium, cerebral blood volume and flow. Intensity and texture features are frequently selected, while anisotropic diffusion coefficient, time to peak and mean time transit mappings seem irrelevant.
4) Conclusion: Multimodal radiomics improve the classification and are consistent with the radiological analysis.
Emerging Value-Based Radiology in the Era of Artificial Intelligence
Abstract
Radiology has a long history of adopting state-of-the-art digital technology to provide better diagnostic services and facilitate advances in image-based therapeutics throughout the healthcare system. The radiology community has been developing diagnostic artificial intelligence (AI) tools over the past 20 years, long before AI became fashionable in the public press. Currently, there are approximately four hundred Food and Drug Administration approved imaging AI products. However, the clinical adoption of these products in radiology has been relatively dismal, indicating that the current technology-push model needs to evolve into the demand-pull model. We will review the current state of AI use in radiology from the perspective of clinical adoption and explore the ways in which AI products will become an ensemble of critically important tools to help radiology transition from volume-based service to value-based healthcare. This transition will create new demands for AI technologies. We contrast the current “technology-push” model with a “demand-pull” model that will aligns technology with user priorities.
We summarize the lessons learned from AI experience over the past twenty years, mainly working with computer-aided detection for breast cancers and lung cancers. The radiology community calls for AI tools that can do more than detection with increasing attention toward higher workflow efficiency and higher productivity of radiologists. Major radiological societies of North America and Europe promulgated the emerging concept of value-based radiology service, an integral part of overall value-based healthcare. The transition to value-based radiology will happen and that higher value will come from the effective use of AI throughout the radiology workflow.
The value-based radiology will need to work with a full range of machine learning tools, including supervised, unsupervised, and reinforcement learning, as well as natural language processing and large language models (e.g., chatbots). The engineering community is rapidly developing many concepts and sophisticated software tools for data orchestration, AI orchestration, and automation orchestration. Current radiology operation has been supported by PACS, a monolithic IT infrastructure of past generations. This system will need to migrate to an intelligence management system to support the new workflow needed for high value radiology.
Prof Cerebrospinal CNS-Leaks: State of the Art in Imaging Diagnosis with Special Focus on Intrathecal MRI Paramagnetic Contrast Agents Procedures
Abstract
MR imaging has proved invaluable in anatomic depiction of the cerebrospinal fluid (CSF) spaces and the surrounding neural and non-neural tissue, although there are still some clinical situations (i.e., cases of CSF-flow alterations, communicating or non-communicating cyst masses bordering CSF pathways, or craniospinal CSF leaks) in which further imaging tests may be required for a definitive diagnosis.
This paper will review the state-of-the-art imaging in these processes, including Radionuclide Cisternography, plain Computed Tomography (CT) and enhanced-CT Cisternography/ myelography, as well as Magnetic Resonance Imaging (MRI) and contrast-enhanced Cisternography/myelography, emphasizing the latest CT and MR imaging refinement advances and proposing tailored specific approaches for two well-established clinical syndromes, namely CSF rhinorrhea and intracranial hypotension syndrome.
Tolerability and Stability of Mask Fixation in Gamma Knife Stereotactic Radiosurgery: Predictors of Treatment Interruptions
Abstract
Background. Frameless fixation with a thermoplastic mask is an alternative to traditional frame-based immobilization for Gamma-Knife stereotactic radiosurgery (SRS) or fractionated stereotactic radiotherapy (FSRT). However, interruptions during beam-on time can significantly prolong treatment delivery, impacting patient experience and unit workflow.
Aim. We investigated clinical and technical predictors of treatment interruptions, and the phases of treatment during which interruptions are most likely to occur.
Methods. Patients undergoing frameless Gamma Knife SRS or FSRT in 2020 were retrospectively reviewed. Clinical parameters were extracted from electronic medical records. Dosimetric and treatment interruption data were obtained from Gamma Knife treatment reports. Univariate and multivariate analyses analyzed technical and clinical predictors of treatment interruptions.
Results. Our cohort included 84 patients receiving 141 fractions encompassing 255 lesions. 49/84 (58.3%) were female, 79/84 (94.0%) had brain metastases, 49/84 (58.3%) were taking dexamethasone and 30/84 (35.7%) used analgesics. 89/106 (84.0%) courses were single fractions. Mean planned beam-on time was 37.1 minutes (range 7.1-118.8 min) versus a total bed time of 64.9 minutes (range 15-252min) per fraction. 64.5% (91/141) of fractions were interrupted at least once; 12/141 fractions were paused 20 times or more, with a maximum 54 pauses. The mean number of pauses per quartile decreased the further the patient proceeded in beam-on time, and patients receiving first lifetime cranial radiation paused more often than during subsequent fractions. At least one pause occurred in 100% of fractions with a planned beam-on time exceeding 60 minutes. Planned beam-on time, number of gating events and high-definition motion management alarms significantly correlated with total number of pauses on multivariate analysis (all p<0.0001); these three factors, along with prep time and number of operator-initiated pauses, predicted total time on the Gamma Knife couch (all p<0.0001). Clinical factors, medication use, and prior SRS/FSRT were not predictive of pauses.
Conclusions. Planned beam-on time, number of gating events and high-definition motion management alarms significantly predicted likelihood of interruptions during frameless Gamma Knife SRS/FSRT. These factors should be considered in selection of immobilization method, especially if exceeding 60 minutes.
